I just roll my eyes at all the biological inspiration stuff. Who cares? If it leads to a useful model family, that’s what matters. It’s not like we’re all chasing after spiking neural networks even though they’re almost certainly closer to what bio brains do.