This code is not doing anything useful:
if (kount2rpm != currentRPM);
As a general comment, your function names 'serloop1' etc are utterly useless and the structure of the code does not seem remotely sensible.
I would expect to find a state machine at the high level dealing with the logic to engage and disengage cruise control. In the 'engaged' mode I'd expect to find code managing the target speed in response to user commands to increment/decrement it, code to determine the actual speed and take action to make the actual speed move towards the target speed. The function name 'getRPM' at least indicates what it does, although it's a blocking function working on a very short timescale and that doesn't seem like a very good approach - I think it would work better to use an asynchronous approach to count revolutions in a period and then calculate the 'current speed' at the end of each period. Blink without delay shows you how to carry out processing at regular intervals, and I'd use that to trigger the calculation at the end of each period. This has the disadvantage that the RPM Hall sensor would only be polled once per loop() rather than being polled intenseively in getROM(). Can you actually afford to poll the RPM Hall sensor in the background? If not, I'd consider counting pulses using an interrupt. Given that the result will be used to drive a mechanical actuator I assume you want to avoid jitter, so I'd suggest applying some smoothing to the calculated speed - an exponential average would probably work well (e.g. smoothed speed equals 90% of previous smoothed speed plus 10% of measured speed).