I doubt that the crystal oscillator was actually jumping from 11 to 12 MHz - it may have looked that way on the scope due to the noise. Adding extra caps sounds like a band-aid - you need to find out where the noise is coming from and arrange your wiring such that noise currents aren't passed through your microprocessor circuitry. Try to visualise where the currents are passing - make sure they don't pass where they're not wanted.

Are you aware of running your power in a star configuration?