They build models to match the observed data. When the models don't match new data they develop new theories and test them against the new data. That's how real modeling is done not the fairy tale that people who like to ignore models promote. Rigging the model is detectable by peer review and would open people to ridicule. The idea that fixing the results is the main mode of computer modeling is ridiculous.

I don't know where you get the idea that distant giants aren't covered by current models either. All it takes is a gravitational transfer of momentum between two gas giants that sends one spiraling inward and one outward. I'm pretty sure I have read of such a prediction which is evidence of good theoretical modeling not "perceptual" modeling.

Regards,
Mark