Not yet reviewed by Pith; the record is open.

We study the constraints imposed by neutrino oscillation experiments on the minimal extensions of the Standard Model (SM) with $n_R$ gauge singlet fermions ("right-handed neutrinos"), that can account for neutrino masses. We consider the most general coupling to SM fields of the new fields, in particular those that break lepton number and we do not assume any a priori hierarchy in the mass parameters. We proceed to analyze these models starting from the lowest level of complexity, defined by the number of extra fermionic degrees of freedom. The simplest choice that has enough free parameters in principle (i.e. two mass differences and two angles) to explain the confirmed solar and atmospheric oscillations corresponds to $n_R=1$. This minimal choice is shown to be excluded by data. The next-to-minimal choice corresponds to $n_R=2$. We perform a systematic study of the full parameter space in the limit of degenerate Majorana masses by requiring that at least two neutrino mass differences correspond to those established by solar and atmospheric oscillations. We identify several types of spectra that can fit long-baseline reactor and accelerator neutrino oscillation data, but fail in explaining solar and/or atmospheric data. The only two solutions that survive are the expected seesaw and quasi-Dirac regions, for which we set lower and upper bounds respectively on the Majorana mass scale. Solar data from neutral current measurements provide essential information to constrain the quasi-Dirac region. The possibility to accommodate the LSND/MiniBoone and reactor anomalies, and the implications for neutrinoless double-beta decay and tritium beta decay are briefly discussed.