I wouldn't claim that philosophy is not relevant, I would claim that our current batch of philosophers are.
I'm a philosopher in as much a sense as it matters to this question, and I don't need others' approval to reach a conclusion. I'm not trying to run a popularity contest, after all.
>What does "one that works" mean?
It means it is both consistent with reality and useful.
>There's no reason why only brain states that have occurred would be relevant.
The reason those ones are relevant is because those are the only ones available for analysis. Those are the brains that determine what "human" currently means. We could evolve into toads in 100B years, but you'd hardly aspire to create a toad brain and call it a good human brain.
>I think it would be uncontroversial to claim that there are an infinity of such hypothetical situations, and each model of morality can deal with them differently, ergo you get an infinity of possible models.
You are confusing the multitude of possible moralities with the multitude of counterfactuals they model. More importantly, you or forgetting that we are only concerned with relevant moralities and relevant counterfactuals, and that we have to make our developments incrementally, and in finite time, which massively constrains the possibilities.
There are an infinite number of ways to skin a cat, but that doesn't mean we can't have a useful notion of the "right way" to do it.