Two ideas that have come to me while running (like a gazelle) round my local park this morning . . .
1) Looking at the map at the start of the game the entrenchment levels for Major Forts - Forts - Fortified Towns - other hexes is 6-4-3-0. What if these were increased a bit? Maybe to 8-6-4-0 or even 10-7-5-0? The entrenchment maximum seems to increase by +1 each time you complete a Trench Warfare research so this would give overall maximums at Trench Warfare Tech 5 of 13-11-9-5 or even 15-12-10-5 instead of 11-9-8-5.
That's one idea anyway. I am not sure it is any good really because I think the main issue is the amount of damage that a single artillery salvo can do - it will definitely De-entrench 1 level, it may kill a defending unit's Strength Point, and it will definitely reduce Readiness and Morale.
So idea 2 is a bit more to the point, I feel, and can either be used in conjunction with idea 1, or be used on its own . . .
2) No strength points of a defending unit can be lost at all from artillery fire if the entrenchment level is 6 or over. So the first point to make is that this idea would not affect the vast majority of hexes in the game as entrenchment level 5 is the maximum they can reach with full Trench Warfare Tech research. But it would effect those few key hexes where entrenchment was between 6 and 11 (in the current game) or at higher entrenchment levels (if any part of idea 1 was in use). That is to say, Major Forts, Forts and Fortified Towns. To simulate the initial breaking up of fortified positions, which would require an initial concerted bombardment, you could also say that artillery salvoes only have a 50% chance of de-entrenching a unit at entrenchment levels of 6 or over.
I think something along these lines might be interesting. I have no idea how to Mod or I would make a "Fortress" Mod myself, but maybe somebody reading this might find it a challenge. The values I have suggested are not necessarily the optimum ones and players with more knowledge of the period than me may be able to improve on them. The other idea I had was to break down the 6 to 11 entrenchment level category into 2 parts (maybe 6-8 and 9-11) and have a differential whereby there was only a 25% chance of de-entrenching at the 9-11 level, but that might be unnecessarily fussy.