Actually one way they can assess previous tectonic formations is studying the rocks that form at mid-oceanic ridges and at "divergent subduction zones", trench areas with seismic activity, were one plate slides under another. Every 1 million years or so the magnetic field of the earth inverts, and this difference affects the polarities of magnetic minerals in cooling molten rock. As a result, moving away from a ridge one notices an alternating pattern of magnetic polarities in the rocks (measured with a magnometer [sp?]).
Here's a site on magnetic striping.
Now using this information, they have a good idea of where plates were at particular points in time.
One thing I notice is that all the evidence your model provides is based on the "continental fit". Sure, it's a good indicator, but not necessarily a reason for the overhaul of tectonic theory alone. Maybe seeing your models will help.
In the mean I found this site (note the immediate graphic), which implies something different to what you've proposed:
http://faculty.valencia.cc.fl.us/jbeeman/A...ic/mesozoic.htm