@NicoleCRust I could be wrong, or behind the literature, but I feel like most of the salient roadblocks in neuro boil down to a few things (of which I'm aware I'm nowhere near qualified to speak on beyond approaching them as mere concepts).

Emergence and nonequilibrium dynamics, and to be more specific the statistical physics behind the stochasticity in neural dynamics and how much of this could even be resolved with our current mathematical suite of models and if we might have to have a few breakthroughs in physics for a paradigm shift for us to start having more pragmatic models. I could be way out of my depth here too so forgive me for any incorrect assumptions but I also think ultimately, we might have to go back to the drawing board with a few things like the molecular dynamics of neural morphogenesis and start looking at all the optimisation constraints that arise from the sheer bump and grind of intracellular mechanics and try to delineate (if possible) how much is genetic, and how much is stochastic brownian motion and if there's a certain gradient of error that's accounted for in our genomic blueprints. I also think that self-organization is something that's selected for by virtue of entropy tending to prefer certain arrangements that faciliate organization as per Jeremy England's paper from 2013 but I'm not sure how much more work's been done on that front.

Basically, as per Stuart Kauffman, "we must delineate the spontaneous sources of order, the self-organized properties of simple and complex systems which provide the inherent order evolution has to work with ab initio and always [..] we must understand which properties of complex living systems confer on the systems their capacities to adapt". I've also wondered if it might be beneficial to have a more dynamic definition of life in a la Schroedinger where we define it as somewhat self-contained systems that replicate and self-organize along lines of entropy gradients. And the interesting thing from there would be.. I'm still not sure of a proper term for this but with respect to xenobots, behavioral convergence, where where various emergent systems (across various degrees of life and "non-life" phenomenon) seem to self-organize along similar lines of function given similar optimization constraints, for instance, how entropic inertia in systems tends to gravitate towards certain 'vacuum basins'(?), regulated by differential gradients and how these phenomena give rise to systematic self-organization. Couple this with how emergence tends to be a fractal phenomenon in nonequilibrium systems.

I realise these seem somewhere far removed from neuro though but I think they're important as far as pinning down the molecular dynamics of the systems from which most neural phenomenon emerge in the first place.