> one of the religions might actually be correct

How would you propose that we set about determining which religion or mythology is "correct"?

> until we have established that religions do not reflect reality, we should at least remain open to that possibility.

The burden of proof is on the religious. If they can offer testable, reproducible proof that a religion makes valid predictions in specific circumstances, and explains how the universe operates, then they should do so. Once they have, we can call it "science".

In the meantime, as Christopher Hitchens said, "that which can be asserted without evidence, can be dismissed without evidence."