ORIGINAL: DrRansom
I'd say that direct control really deals with an AI problem and the real inaccurate element of CMANO simulation: the Perfect Data Link (tm).
Direct control makes sense in a situation where the unit AI does not behave as you'd expect but is very natural. An example of this is a fighter wing being vectored in to ambush enemy aircraft by an AWACs. An air superiority mission doesn't produce the realistic behavior. Now, you could just use the air superiority mission, but the unit behavior doesn't fully exploit the tactical situation.
In my opinion, the real issue is the perfect data awareness for all units. This produces unit behavior which is more optimal than reality. Again, return to the issue of fighter interception. The above scenario (AWACS + Fighters) exists in real life. In CMANO, the same behavior can be achieved with any ground based radar and fighters. So, a tactical SAM system, e.g. SA-6, can serve as a GCI for interceptors.
Now, that can be fixed by playing with the multiple sides, but perhaps there is another solution? Maybe an 'acquisition time' could be added to different units, to reflect the process required to detect a target. This acquisition time can be adjusted by having proper GCI / AWACs units in the region.
In a way, there would be a new type of datalink which affects the performance of a new parameter ('acquisition time'). Perhaps that can introduce some of the friction that would naturally exist in combat?
We dislike Borg hive-minds too. Stick around.

