We have re-examined gas phase silicon chemistry in dense molecular clouds. We suggest that the fractional abundance of SiO in both cold and warm quiescent conditions, which is observed (non-detection limits ~10^-12-10^-11) to be much lower than predicted (~10^-8), may be accounted for without the need to invoke extreme depletion or a particular critical temperature and density. We reconsider the standard grain mantle model for silicon and the availability of silicon following hydrogenation on grain surfaces. We also consider the gas phase impact of high neutral carbon abundance detected within dense clouds in recent years. Applied also to more active star-forming regions, our results in this and a previous paper are consistent with recent suggestions that shock disruption of grains need not be the origin of all observed SiO.