Recent and not-so-recent critiques of the widely used magnetohydrodynamics (MHD) code, ZEUS-3D, challenge its reliability and efficiency suggesting that its MHD algorithm is capable of "significant errors" in some simple one-dimensional shock-tube problems. I show that these concerns are either inapplicable in multi-dimensional astrophysical applications, or result from a misuse of the code rather than "flaws" in its design. I also describe a few multi-dimensional test problems including one for super-Alfvenic turbulence, and highlight some recent innovations and improvements to the code now available online.