I'm not knowledgeable about the efficacy of the masks themselves, but I suspect there's a forest-for-the-trees issue, here. There are at least a few ways to think about whether masks (broadly, or any specific mask) "work"...
- Whether wearing a mask makes it less likely someone will pick up an infection in constant conditions (direct mask efficacy, reduced touching, etc.)
- Whether wearing a mask makes it less likely someone will transmit an infection in constant conditions.
- Whether wearing a mask leads infected people to spend more time around others.
- Whether wearing a mask leads uninfected people to spend more time around others.
- Whether the way masks end up getting allocated in the event of a shortage promotes or inhibits spread at population scale (for example, if masks ended up overallocated towards people who're so concerned that they stay home anyways and away from people who don't have the ability to).