Strength losses too low
The casualties in Poland was around 200,000 in one month, and then there were about 700,000 captured out of a force of about a million (20% casualty, 70% POW). If the average brigade spent 100 hours in contact, then there should be about .2% losses per hour, or 6 per hour per brigade.
Keep in mind that for the Russian army in the modern era, 10% casualties in an exercise wasn't too abnormal. In exercises, US and British para divisions experience about 5% casualties per year in jump exercises (typically 5 jumps a year). Combat, poorer training, tactics, equipment, nutrition and conditioning, medical response, etc imply WW2 rates should be significantly higher. Paradrops should start with perhaps a 3% loss at touchdown.
Engineers shouldn't increase defense, they should increase fortification. Minefields, obstacles, entrenchments, etc already has one game mechanic. This is important because the counter to defensive engineers is offensive engineers, which undo these changes to the battlefield.
Headquarters brigades should contain all elements necessary to maintain communications. Signal brigades do exist at the corps level, and it would make sense that signal would increase C2 range, and intelligence/counterintelligence functions. No other unit type however makes sense for this.
Suppression should propagate over whatever borders troops can cross. This is abstracting battalion or company level units spreading out, and they should have less restrictions on occupation movement than they would on combat movement.