Well certainly the more people who do it, the better.
Sure, but when you look at how facors like this and social distancing affect transmission using a individualised SIR model the effectiveness drops massively when even a relatively small proportion don't do it. If take up is only 50% it is likely to have very little effect on the curve- which is really the main point.
Masks are not tremendously effective on their own, but they seem to be working well in places like China and Singapore where high compliance with a range of measures (coupled with tons of teting and contact tracing) is enabling them to function fairly normally.