Compression and limiting is essentially the same thing, the terminology/difference comes in that limiting is when you have high ratios (typically 10:1 or more), and very fast attack and release settings (on the order of milliseconds). So it's likely the "too much compression" with fast/high settings, but I also do not know all the export settings possible and I am not a video person.
What I am hearing is that the audio gets compressed/limited into the RMS portion of the sound, rather than just cutting down the peaks. What that means for how it sounds, is that the body of all voices/music is a constant volume, and nuances in voices or other small fluctuations are not any louder at all than the body of the sound/voice. This is why it gets fatiguing to me. If you back off on the compressor by a few DB just so that it's not always pinned, it will not limit the body/RMS sound all of the time (of course RMS is calculated over an average of time, so it can be calculated differently...but I'm certainly hearing it is compressing everything), and will actually lead to more nuances coming through. Your videos may end up a couple of dB quieter than they are now (peak will be the same, but RMS will be slightly lower), but SpinTV videos are already very very loud so you can afford to lose some dB.