So, if I understand correctly, the upshot of all this technical jiggery-pokery is that aelf is not guaranteed to get the religion, despite having an oasis. On the other hand, if there are no AI civs that start with mysticism, or if all of those that do have it go for the other religion, then he's guaranteed to get it (with the oasis shaving off a turn or two, so as to move more quickly onto the worker techs).
Overall, then, we're either:
1) getting into seriously complicated probability calculations, to figure out the chances of winning this particular race
2) playing aggressively and taking the risk of going for a religion
3) playing conservatively and leaving the early religions to the AI
Personally, I would say hang the exact probabilities; you've got a half-decent chance of grabbing an early religion, and that's enough to work with; imo, the key questions are:
a) What is the opportunity cost of going for a religion at this time?
b) Which religion represents the better/safer choice?
c) Would losing the race be such a disaster that it's not worth taking the risk?
This being immortal difficulty, perhaps the conservative option would be more sensible; concentrate on worker techs and getting the food and hammers rolling in, so as to get those settlers out asap.
Nevertheless, I'm always inclined to bold moves in situations like this, so I'd say go for a religion. Hinduism looks the better bet to me; from my experience you've got more chance of getting it and, should you fail, at least you'll have picked up a prerequisite for the GL. There's no guarantee that any of the AI civs on your continent will land an early religion, in which case having your own one would be a huge advantage. And the happiness bonus would also be most helpful, given the low happy cap at this level.
I reckon it's a risk worth taking.