A scalar field model for explaining the anomalous acceleration and light deflection at galactic and cluster scales, without further dark matter, is presented. It is formulated in a scale covariant scalar tensor theory of gravity in the framework of integrable Weyl geometry and presupposes two different phases for the scalar field, like the superfluid approach of Berezhiani/Khoury. In low acceleration regimes of static gravitational fields (in the Einstein frame) with accordingly low values of the scalar field gradient, the scalar field Lagrangian combines a cubic kinetic term similar to the ``a-quadratic'' Lagrangian used in the first covariant generalization of MOND (RAQUAL) (Bekenstein/Milgrom:1984) and a second order derivative term introduced by Novello et al. in the context of a Weyl geometric approach to cosmology (Novello/Oliveira_et al:1993, Oliveira/Salim/Sautu:1997). In varying with regard to $\phi$ the latter is variationally equivalent to a first order expression. The scalar field equation thus remains of order two. In the Einstein frame it assumes the form of a covariant generalization of the Milgrom equation known from the classical MOND approach. It implies a corresponding ``non-metrical'' contribution to the acceleration of free fall trajectories. The second order derivative term of the Lagrangian leads to a non-negligible contribution to the energy momentum tensor and an {\em add-on to the light deflection potential} in beautiful agreement with the dynamics of low velocity trajectories. -- In higher sectional curvature regions, respectively for higher accelerations in static fields, the scalar field Lagrangian consists of a Jordan-Brans-Dicke term with sufficiently high value of the JBD-constant to satisfy empirical constraints. Here the dynamics agrees effectively with the one of Einstein gravity.

A model for the dark halos of galaxy clusters, based on the Weyl geometric scalar tensor theory of gravity (WST) with a MOND-like approximation, is proposed. It is uniquely determined by the baryonic mass distribution of hot gas and stars. A first heuristic check against empirical data for 19 clusters (2 of which are outliers), taken from the literature, shows encouraging results. Modulo a caveat resulting from different background theories (Einstein gravity plus $\Lambda CDM$ versus WST), the total mass for 15 of the outlier reduced ensemble of 17 clusters seems to be predicted correctly (in the sense of overlapping $1\,\sigma$ error intervals).

A scale invariant, Weyl geometric, Lagrangian approach to cosmology is explored, with a a scalar field phi of (scale) weight -1 as a crucial ingredient besides classical matter \cite{Tann:Diss,Drechsler:Higgs}. For a particularly simple class of Weyl geometric models (called {\em Einstein-Weyl universes}) the Klein-Gordon equation for phi is explicitly solvable. In this case the energy-stress tensor of the scalar field consists of a vacuum-like term Lambda g_{mu nu} with variable coefficient Lambda, depending on matter density and spacetime geometry, and of a dark matter like term. Under certain assumptions on parameter constellations, the energy-stress tensor of the phi-field keeps Einstein-Weyl universes in locally stable equilibrium. A short glance at observational data, in particular supernovae Ia (Riess ea 2007), shows interesting empirical properties of these models.

The often quoted equ. (15) in (Anderson_ea:2002) defines the observed Pioneer anomaly $a_P$ such that literal reading suggests $\nu_{obs} < \nu_{mod$ ($\nu _{obs}$ the observed (Doppler) frequency, $\nu_{mod}$ the one expected from a relativistic orbit model of the spacecraft). Contextualized reading shows, however, that (15) is meant the other way round, $\nu_{obs} > \nu_{mod}$. Version 1 to 4 of this paper built upon the first reading and led to a correction term for frequency shifts of wrong sign. Thus the proposal therein does not resolve the Pioneer anomaly, although it gives a correction term for frequencies which has the correct absolute value.

A conservative extension of general relativity by integrable Weyl geometry is formulated, and a new class of cosmological models ({\em Weyl universes}) is introduced and studied. A short discussion of how these new models behave in the light of observational evidence follows. It turns out that the subclass of Weyl universes with positive ex-ante curvature of spatial sections ({\em Einstein-Weyl universes}) forms an empirically contentful conceptual frame for cosmology. Recent high precision data (supernovae Ia, quasar frequencies, microwave background) are well represented in this type of Weyl geometry and several of the anomalies of the standard approach are avoided or become meaningless.

Already the simplest examples of Weyl geometry, the static space-time models of general relativity modified by an additional time-homogeneous Weylian length connection lead to beautiful cosmological models (Weyl universes) . The magnitude-redshift relation of recent supernovae Ia measurements is in perfect agreement with the prediction of decrease of energy flux in the Weyl models. These data allow to estimate the (ex-ante) spacelike curvature of Weyl universes. Quasar frequency data from the SDSS provide strong evidence of a positive ex-ante curvature. Thus an Einstein-Weyl universe, i.e., an Einstein universe endowed with a Weylian length connection, is in good agreement with supernovae and quasar data. The relative mass-energy density with respect to the critical density of the standard approach, and the relative contribution of the ``vacuum term'' are time-independent in Weyl gauge. Thus the time-evolution anomaly of vacuum energy does not arise. The intervals given in the literature for the dynamically determined matter density parameter are consistent with other data in the Weyl geometric approach. An explanation of the cosmic microwave background by a maximal entropy state of the quantized Maxwell equation, according to a proposal by I.E. Segal for the conformal Einstein universe, translates directly to the Weyl approach. It leads to quantitative estimations for the temperature of the Planck radiation between 2 and 5.3 deg K. Moreover, the anisotropy angles, expected to arise in an Einstein-Weyl universe agrees with the results of multipole measurements of CMB anisotropies.