I think they'd basically need a practically loss-less energy conversion of electricity into hydrogen for it to compete with direct electricity into cars. There is a roughly 10% loss when charging a battery, so for hydrogen conversion to compete with that, it would need to be improved from a 30%-ish efficiency to 95%+, because it would also take energy to move the hydrogen around afterwards. On top of that, the fuel cell efficiency is only at most 45% efficient as well. Both these would have to be improved immensely before it reaches the level of efficiency all electric cars already have today, and it's not like that tech will stand still for the next years either.
A hydrogen car isn't more complex than all electric in some ways, it is more complex in every way. The driveline of a hydrogen car is basically exactly the same as an electric car. It has a cache battery (to make up for the fact that a fuel cell can't generate enough power to give a car even 100 hp, so it needs help during acceleration), then it has electric motors, and a solution for harvesting brake energy back into the battery. So far, it has the same complexity as an all electric.
However, now you also add a hydrogen tank, a fuel cell, and a unit to safely decompress hydrogen from 200 bars to an usable pressure, which consumes a lot of heat, effectively freezing the components, which makes them weaker unless you immediately supply more heat to make up for it. Then the fuel cell also gets clogged up with junk over time, which lowers its efficiency and requires you to service the car more often.
It can never be less complex than an EV because it is basically an EV with 50 extra things attached. Many people go for electric vehicles despite their higher cost because they expect this extra cost to be made back through fewer fuel and service expenses. Hydrogen is never going to get as cheap to tank up with as electricity from your house will be, and I definitely imagine a hydrogen car requiring a lot more servicing than an EV.