An underlying question is whether turning a stage off in the design window should eliminate that stage from the launched vehicle. The primary concern in implementing that approach is what to do is the user turns off the primary or an intermediate stage.

With that question in mind, using the Three-stage rocket example, the current implementation appears contradictory. As can be seen below, if the second booster is turned off, the first simulation removes the corresponding motor, but none of the other results change (the other simulations are unaffected). This is the same functionality if the first booster, and the primary stage are successively turned off.

So, the first simulation is giving the user the impression that the second booster stage that is turned off is not part of the launch vehicle, even though the launch results include it. And the same is true when the first booster is turned off, and even the primary stage.

If turning off a stage should eliminate that stage from the launch vehicle, then changes to implement that approach need to be done. Or, if turning off a stage should not eliminate that stage from the launch vehicle, then changes to implement that approach need done. But, the behavior should be consistent across all flight configurations.

Boosters are behaving similarly.