We briefly review a theory for the cuprates that has been recently proposed based on the movement and interaction of holes in antiferromagnetic (AF) backgrounds. A robust peak in the hole density of states (DOS) is crucial to produce a large critical temperature once a source of hole attraction is identified. The predictions of this scenario are compared with experiments. The stability of the calculations after modifying some of the original assumptions is addressed. We find that if the dispersion is changed from an antiferromagnetic band at half-filling to a tight binding $cosk_x + cosk_y$ narrow band at $<n> =0.87$, the main conclusions of the approach remain basically the same i.e. superconductivity appears in the $d_{x^2 - y^2}$-channel and $T_c$ is enhanced by a large DOS. The main features distinguishing these ideas from more standard theories based on antiferromagnetic correlations are here discussed.

We review the results of an extensive investigation of photoemission spectral weight using electronic models for the high-Tc superconductors. Here we show that some recently reported unusual features of the cuprates namely the presence of (i) flat bands, (ii) small quasiparticle bandwidths, and (iii) antiferromagnetically induced weight, have all a natural explanation within the context of holes moving in the presence of robust antiferromagnetic correlations. Introducing interactions among the hole carriers, a model is constructed which has ${\rm d_{x^2 - y^2}}$ superconductivity, an optimal doping of $\sim 15\%$ (caused by the presence of a large density of states at the top of the valence band), and a critical temperature $\sim 100K$.

Recent angle resolved photoemission (ARPES) results for the insulating cuprate ${\rm Sr_2 Cu O_2 Cl_2}$ have provided the first experimental data which can be directly compared to the (theoretically) well--studied problem of a single hole propagating in an antiferromagnet. The ARPES results reported a small bandwidth, providing evidence for the existence of strong correlations in the cuprates. However, in the same experiment some discrepancies with the familiar 2D ${\rm t-J}$ model were also observed. Here we discuss a comparison between the ARPES results and the quasiparticle dispersion of both (i) the ${\rm t-t'-J}$ Hamiltonian and (ii) the three--band Hubbard model in the strong--coupling limit. Both model Hamiltonians show that the experimentally observed one--hole band structure can be approximately reproduced using reasonable values for ${\rm t'}$, or the direct oxygen hopping amplitude ${\rm t_{pp}}$.

Numerical and analytical studies of several models of correlated electrons are discussed. Based on exact diagonalization and variational Monte Carlo techniques, we have found strong indications that the two dimensional t-J model superconducts near phase separation in the regime of quarter-filling density, in agreement with previous results reported by Dagotto and Riera (Phys. Rev. Letters 70, 682 (1993)). At this density the dominant channel is d_{x^2-y^2}, but a novel transition to s-wave superconductivity is observed decreasing the electronic density. In addition, the one band t-U-V model has also been studied using the mean-field approximation that accurately described the spin density wave phase of the repulsive Hubbard model at half-filling. Finally, the two band Hubbard model on a chain is also analyzed. Superconducting correlations near phase separation exist in this model, as it occurs in the t-J model. Based on these nontrivial examples it is $conjectured$ that electronic models tend to have superconducting phases in the vicinity of phase separation.Reciprocally, if it is established that a model that does not phase separate, then its chances of presenting a superconducting phase are considerably reduced.