Overhead power appears to be the effective solution for the branch in general, and if it was a simple branch like Henley then they could probably have all been wired along with the GWML and be running with Cl387s today.
Whilst wiring to Bourne End would be straightforward, the Marlow end of the route presents issues, in the form of the short platform at Bourne End, and based on posts above the safety concerns with clearance at crossings. A self propelled two car train therefore remains the solution to get to the riverside terminus. Usage seems to show that only one train is needed on the whole route off peak, so something needs to be able to do both, which leads to the "need" being something like a two car EMU, with alternative power for the Marlow stub. The desire to lose diesel then leads us batteries to enable them to reach Marlow, presumably charging on the run down to Bourne End.
Off peak, such a train operating on the branch would spend forty minutes in the hour under the wires, which should with modern developments be enough to charge the batteries for the twenty minutes it takes to get to Marlow and back. The challenge will be at peak times, where a set is presently orphaned at the Marlow end. At best it can sit at Bourne End for ten minutes on each cycle - would this be enough to power a 20 minute round trip? One can assume that if wired a 387 can run the Maidenhead to Bourne End part of the shuttle at peak times.
Its clearly not within the realms of EMU development to design and build 2 car overhead powered unit (they managed it sixty years ago) but there has never seemed to have been a need for any more since, but no reason why it could not be done, coupled with battery developments, but with so few units required is the major hurdle really the cost of developing such a small microfleet, effectively two trains, for a utilisation of 50%?