Next point is about the diversity of game play experience during campaign. From my point of view, campaign might be repetitive and eventually become boring with current system. In this case, 'repetitive' means not about maps and units, but about mission type and mission targets. With current system, all missions in campaign will be very similar, frontal attack/defense game against the opponent on the otherside of tactical map.
Of course, tactical map itself and opponent forces will be changed, but the core contents will be the same, 1:1 battle against AI on linear frontline. During the campaign, player will be only allowed to attack the enemy on the other side of the map, or defend from the enemy coming from other side of the map. Campaign might become continuous collection of similar battles over and over again.
So... how about introduce more mission types and more mission goals? In current system, players cannot enjoy something like "evacuation" mission of WEE. Or fight against the chasing opponent from the rear, while retreat/march to other side of the map. Or protect and evacuate civilians and their vehicles as much as possible from city during last phase of sudden retreat / evacuation. Or defense against multiple enemy AIs assaulting from all directions until reinforcement arrives... I guess most of them might be possible to introduce with current system, by tweaking of VP locations or slightly modify VP-oriented AI or etc... But I don't know about the source code of this game, so I maybe totally wrong here. Civilian AIs looks not easy to me, though.
Then it would be possible to categorize missions, something like "offensive mission types", "defensive mission types", and "neutral mission types". Then, introduce assault / retreat factor to the campaign generator, which can be set by user, or randomly generated based on the direction of the waypoint from the campaign map, or campaign difficulty setting. Depending on random numbers, location of tactical battle, campaign progress or campaign score, and morale or other factors, maybe it would be possible for campaign generators to initiate such "unique" offensive or defensive missions in the middle of automated campaign. I wish those will make campaign much more dynamic.
This is just my brainstorming ideas, so please don't be burdened. But there might be very very interesting missions which players can immerse themselves to the game. I think such contents will make this game far richer than before.
The current Campaign Generator system is mostly about what happens before a battle and after a battle. What you are suggesting are new battle types, and even if more mission types can be added in the future, that was not the emphasis this time. The main idea was to provide a context and continuum to the battles, using the existing battle system.
ps) How about showing the movement of friendly and enemy army units from operational layer and campaign map? Players cannot control the units from operational layer, as a commander of one of the units shown in campaign map. But just showing would be still very helpful.
There have been different ideas about showing formations, support, artillery, or whatever on the campaign map, but we haven't been able to settle with anything yet.
Then, introduce operational layer AI, to strategically moving / directing the forces under command. (I guess operational layer AI might not that difficult when compared to tactical layer AIs?) Then, allow campaign generator to create unique missions based on campaign progress shown from operational layer. In this case, many unique missions, like battle under surrounded situation, or battle against rear enemy during retreat... would much much more sense and appeal to players.
In this case, showing the unit movement and campaign status from operational layer to players have additional advantage, increase the immersion and concentration of players, by showing players the reasonable motivation of their situation and their battle.
A simple operational level game is probably easier to implement than the complex tactical level game that we have, but still, it would be like making another game. In this project we must carefully weigh what's the best use of our time and resources.