I know 't Hooft has done some work in this area during the last few years, but I don't think he (or anyone) is even close to finding a hidden variable theory that can predict the outcome of experiments as accurately as QED.
No new additions to the standard theory (QM) itself. As to local hidden variable theories, which are forbidden from agreeing with QM per Bell's Theorem: There are a few diehards out there who cling to the idea of a stochastic theory which has local hidden variables. Their thinking is that the standard cos^2(theta) relationship observed in experiments is an illusion, that there are experimental problems which "bias" the results. At this time, there is no seriously considered options for local hidden variable theories. There is a non-local candidate, however: Bohmian Mechanics.