A number of misunderstandings about modeling the apparent accelerated expansion of the Universe, and about the `weak singularity' are clarified: 1. Of the five definitions of the deceleration parameter given by Hirata and Seljak (HS), only $q_1$ is a correct invariant measure of acceleration/deceleration of expansion. The $q_3$ and $q_4$ are unrelated to acceleration in an inhomogeneous model. 2. The averaging over directions involved in the definition of $q_4$ does not correspond to what is done in observational astronomy. 3. HS's equation (38) connecting $q_4$ to the flow invariants gives self-contradictory results when applied at the centre of symmetry of the Lema\^{\i}tre-Tolman (L-T) model. The intermediate equation (31) that determines $q_{3'}$ is correct, but approximate, so it cannot be used for determining the sign of the deceleration parameter. Even so, at the centre of symmetry of the L-T model, it puts no limitation on the sign of $q_{3'}(0)$. 4. The `weak singularity' of Vanderveld {\it et al.} is a conical profile of mass density at the centre - a perfectly acceptable configuration. 5. The so-called `critical point' in the equations of the `inverse problem' for a central observer in an L-T model is a manifestation of the apparent horizon - a common property of the past light cones in zero-lambda L-T models, perfectly manageable if the equations are correctly integrated.

The currently available cosmological data yield, as a most striking result, that the expansion rate of the universe seems to be increasing at late times, contrary to the standard (zero cosmological constant) FLRW prediction. The usual explanation for this discrepancy is that a new component of the energy density of the universe, known as dark energy, dominates this recent evolution. Since the existence of such a new component would have a revolutionary impact on our understanding of the fundamental laws of physics, we think important to check other interpretations. We have therefore shown that the SNIa observations could be reproduced by the effect of inhomogeneities. This idea has been further developed by different teams, and enlarged to other cosmological data. We will give here a review of the results of these works and the prospects for future developments.

Standard cosmology is constructed upon the (generally implicit) assumption of the ``large scale'' homogeneity of our Universe. Now, structures are observed at scales which become larger and larger as the observational distances increase. However, the homogeneous Friedmann-Lemaitre-Robertson-Walker model remains a cosmological paradigm, and Friedmannian relations are usually used to work out the model, implicitly assuming that the homogeneity assumption is valid at the studied scales. This has been the way the concordance (Lambda-CDM) model came out from the analyses of the SNIa data. Since this model implies the validity of the Friedmannian magnitude-redshift relation at the range of redshifts spaned by the supernovae surveys, it has been proposed some years ago a very simple test of this possible validity, provided the SNIa should be confirmed as standard candles. Since we can hope that the redshifts of the supernovae which will be observed in the years to come will allow such a test to be performed, it would be interesting to find inside the SNIa community some collaborators ready to perform the test.

The structure of the large scale distribution of the galaxies have been widely studied since the publication of the first catalogs. Since large redshift samples are available, their analyses seem to show fractal correlations up to the observational limits. The value of the fractal dimension(s) calculated by different authors have become the object of a large debate, as have been the value of the expected transition from fractality to a possible large scale homogeneity. Moreover, some authors have proposed that different scaling regimes might be discerned at different lenght scales. To go further on into this issue, we have applied the correlation integral method to the wider sample currently available. We therefore obtain a fractal dimension of the galaxy distribution which seems to vary by steps whose width might be related to the organization hierarchy observed for the galaxies. This result could explain some of the previous results obtained by other authors from the analyses of less complete catalogs and maybe reconcile their apparent discrepancy. However, the method applied here needs to be further checked, since it produces odd fluctuations at each transition scale, which need to be thoroughly explained.

The existence of stars and galaxies reqires cosmological models with an inhomogeneous matter and radiation distribution. But in these models the initial singularity surface t_0(r) is in general homogeneous (independent of r). In this second paper of a series devoted to an inhomogeneous Big Bang singularity, we investigate the cosmic microwave background radiation (CMBR) dipole. A special Tolman-Bondi Universe is used to study the effect of a Big-Bang singularity, depending linearly on r, on the CMBR anisotropy. It is shown that, for an observer located off the ``center'' of this Universe (r=0), the parameters of the model can be tuned so as to reproduce, with a good approximation, the dipole and the quadrupole moments of the CMBR anisotropy observed in recent experiments. If the dipole should prove cosmological, a slight delaying of the Big-Bang over spatial coordinates would thus be a good candidate for its interpretation.

One of the main drawbacks of standard cosmology, known as the horizon problem, was until now thought to be only solvable in an inflationary scenario. A delayed Big-Bang in an inhomogeneous universe is shown to solve this problem while leaving unimpaired the main successful features of the standard model.