You can have (many!) things which you hold to be important and work to bring about, but which you do not spend every plausibly-available resource on.

What about the argument from marginal effectiveness? I.e. unless the best thing for you to work on is so small that your contribution reduces its marginal effectiveness below that of the second-best thing, you should devote all of your resources to the best thing.

I don't myself act on the conclusion, but I also don't see a flaw in the argument.