You are reasoning like if not 100% of engineers but "only" 99.999% work on the priority projects then those priority projects will fail. That's not the case.

Moreover, there is always the need to continue to explore and diversify research and development. It allows to spread the risks (even if modestly, given that most resources go to the priority projects) and discover the necessary paths to go ahead (which is required even for the critical paths for the priority projects).