An early excuse was that GW was using the new table size divided into quarters for smaller games and expecting people to slide those crappy paper sheets together as mats, due to being some standardized size, but A1 and A2 paper sizes are too big for that as well(the ANSI paper sizes don't fit either).
This is what I heard as well, but there doesn't seem to be any clear cut answer. I agree that the FLG table made sense on paper, but it is kind of wild to think that no one maybe play tested it and realized that they had no where for accessories or drinks. Maybe they assume that people will use a different table away from the board for all that? I don't know who plays like that, but if you require an extra table for your table it defeats the purpose. At that point you may as well invest in one of those fancy gaming dinner tables people use for DnD and board games. I've seen people use pool tables which, conveniently, is typically 44" across, and 88" long so plenty of space in the center for the board and room on the sides for causalities, dice, etc. That can be a really expensive choice for a table, but the upside is its also a pool table.
With the current table size and deployment zone sizes, I don't think cutting back shooting to even mostly cap at 24" would resolve the problem as that's still entirely across NML from one DZ into another, and much longer than a lot of melee threat ranges outside of things like bloodcrushers and jump pack marines. Proper LoS rules would fix a lot of the leapfrogging issues(and make horde armies more playable), as one guy having a sword sticking out from behind a wall doesn't mean the entire unit can get wiped and you only lose 1 model(and using silhouettes or determined volumes off of base size eliminates model overhang being a problem for LoS and cover too, GW even acknowledged this problem recent-ish allowing models to squeeze through terrain gaps if their base would fit but the model wouldn't so long as the move isn't ended in the space where it can't fit(so big models with wings and shit like that).
Cutting ranges would not solve the problem, but no one solution will. I think the combo of AA, 4'x6' boards, and halved ranges will resolve probably most of the big issues. Even if you can shoot across NML, at 24" you would have to be right on the line at the start of the game, and shooting perfectly straight down the board. Assuming there is even some obscuring cover in between this still improves survivability by leaps and bounds.
The other thing OPR does which doesn't translate to 40k is that charging is guaranteed rather than a roll. In 40k 7" is considered the danger zone for melee units, because anything more than that is improbable. In OPR if it's in 12", you're getting charged. I think with these changes we can start to remove some of the artificial fail states built into the game like rolling to charge, or rolling to advance. If we don't eliminate rolling I think we could at least give some kind of minimum to movement. Say you failed a charge, maybe instead of skipping your turn you at least got a 3" surge move? I generally am against any rule that says you have a random chance of literally nothing happens.
I am mixed about changing the LoS rules. On one hand the ruling is very clean. There's no arguing about whether this is considered 50% of the model, or if my commander's pinky toe counts as a valid target for the ass end of a Leman Russ's exhaust pipe. So long as you can connect the dots, it counts. This makes it so easy to LoS that most people just need a quick glance and assume there's LoS without fussing around with lasers and string. On the other hand it makes kitbashing, and modeling a double edged sword. I could add all sorts of bits to my models or make them lean as far forward as possible to give me the most reach. Or it could mean that my centerpiece model becomes a really easy target because how the hell am I supposed to stand behind a ruin with this wings out? Adding a check for valid targets, and restricting LoS will make the game crunchier, and it means that there's a lot more work that will have to go into the shooting phase. You would assume that good sportsmanship means people are generous and wont try to argue about whether or not their tank's gun can draw a line to an imaginary cylinder above the target's base. I think reality is this will cause arguments and we have to decide whether or not the trade off is worth it.
I wish we could use an entirely 2D LoS model that only uses the base itself and lines across the board to determine LoS. It's simple and extremely easy to measure with little room for edge cases. Unfortunately this doesn't account for any kind of 3d space such at shooting from above, or taller models looking over walls. I know the game is like 90% played 2D anyways, but it doesn't work as well as I wish it did. Maybe we should allow units to block LoS around their perimeter? It could add an extra role for some units to act as a phalanx moving up the board.