Development and results of the MHD description of stellarators over the past half-decade are reviewed. Major events were the acceleration of convergence of those equilibrium codes which assume nested surfaces, the advent of equilibrium codes which do not a priori assume nested surfaces, and the advent of stability codes for nonlocal modes. Major results having these developments as prerequisites were that stellarators can definitely be optimized, can have nicely nested surfaces at finite beta at least in a limited sense, and can be completely ideal-MHD stable at elevated beta values.