The so called "Dutch Roll",can easily be designed out of our models.
Since many of our models are copies of a full scale subjects, or aproximations of em, we have relatively the same moment arm relationships, etc..
IF-- you shorten the span , the problem is reduced .
IF you increase the tail volume the problem is reduced -sometimes.
IF you reduce drag in the forward portions of the airframe and increase it in the aft sections - the problem is reduced.
IF you could take a simple arrow and slide the feathers forward along the shaft - you should get into this same situation at some point.
Realistically, the problem in the models appears to be (to me), a situation where the dynamic forces occurring in each wing panel , are more than the tail group can easily stabilize.
All of my own fiddling bears this out.
Sorry - my engineering speak isn't working --but I am certain someone has a formula (similac?) for all of this .