For a direct answer, this[1] is a good starting point discussing the difficulties with hydrogen (as a medium generally).
Hydrogen fuel cell vs combustion isn’t a settled issue. Combustion is more familiar but Carnot Efficiency is going to reduce range (even further). Fuel cells are more efficient but more expensive. Anything with hydrogen needs to worry about embrittlement. It’s also very difficult to contain so the GWP of 11.6 shouldn’t be overlooked; methane is much easier to manage and still our natural gas infrastructure leaks all over the place.
The electrification of industrial processes have allowed for increasingly-cheaper electric motors to operate efficiently at any size, from toy RC cars to massive (diesel) electric trains.
Even barring any energy density breakthroughs, electrification with batteries for light duty applications seems inevitable. Having a gigantic battery (compared to residential electrical use) as backup is quite useful. Virtual power plants [2] and stationary deployment of secondhand battery cells should lessen the substantial cost of vehicle batteries — we’re at least a decade away from EV trade-ins being common outside of Norway (and maybe China).
For heavy duty applications, we have an existing solution that works fairly well. We’re just going to end up reinventing catenaries for this era [2]. The operational challenges of maintaining catenary wire systems (especially with shared electric right of ways) is just easier than overcoming the chemistry. That is of course, unless we get in our own way [4].
—-
[1] https://www.thechemicalengineer.com/features/the-unbearable-....
[2] https://liftoff.energy.gov/vpp/
[3] https://www.carscoops.com/2021/10/germany-is-already-testing...
[4]
https://www.gao.gov/products/gao-08-347r