Skanvak, this is the way to go. Most of the time GB/Fr could skip their declaration of movement phase, so this seperate select when to move phase would not be a factor 95% of the time (likely best to let GB and France determine when they move both during their regular reinforcement phase and during a seperate phase, so there would be no need for a move declaration phase 98% of the time).
There would also have to be a few minor limitations with sim reinforcement to prevent two players from adding corps or garrisons to the same location and going over the corps per space or garrison limits. These might be, 1) you can place anything, anywhere you want in your home nation/free states and conquoured states. But 2) outside of that, you could only add factors to existing corps and depots, or add new corps counters to areas where you have depots, or under your "control", as only one player can control one area at a time (of course this assumes the above areas are in supply). Not a big limitation.
Marshall, I think many PBEM's have learned that to have a good PBEM, we don't need to exactly replicate EIA rules to the letter. Anyways in the old board game, almost all groups treated reinforcement as simultaneous anyways, except for the announcement of who moved when. It might not have been the rules, but it was the way the game was played.
Marshall I think you will less get slammed if you implement a TCP/IP game really true to EIA, FIRST.
Then all modification for PBEM will seem more legitimate (by the way don't forget to make it optional).
As for the order.
Everyone does the simultaneous reinforcement phase,
Add a simultenous French/England move order decision (again England and France can decide simultaneously)
goes to the move order.
But again, implement a true EiA with all the correct phase for TCP/IP first then revise the PBEM, not the otherway round. Doing the PBEM stuff AFTER will give a lot legitimacy to all simplification. If you begin by the PBEM you might angry some players and most will be afraisd the TCP/IP will be like the PBEM, therefore limiting all attempt to simplify the PBEM routine.