As a shade tree mechanic myself I was disappointed to see that your experiment failed. Personally I have limited confidence in chemical bonding.
My first experiment at changing the window was an attempt to glue the new plastic in place. I cut the old window out, leaving about an inch all the way around then glued the new piece in using the inch of overlap and VLB glue. Unfortunately the glue would not adhere to the old plastic (it worked on attaching 2 pieces of the new material together).
So I changed to mechanical attachment. I sewed the new material in place by hand, with the roof in the car using the same 1 inch band I had tried to glue to. Basically I placed the new window, with a sealant between the two layers and attached it with a few rivets in the middle at the top and bottom and the two sides, just to hold it in place. I then used a soldering gun with a small tip to make holes through the two layers of plastic, pulling it tight as I worked my way around. I then sewed the two pieces together by passing the thread through the holes. I did this alone so it took some time. Once the sewing finished I sealed the joints and the holes made for the thread (I think I used the VLB for this), then ran a piece of insulation foam around the perimeter to hide the sewing and glue.
Despited the ripples (which almost disappeared with time) I counted this as a success. I could see through the window and it was waterproof. Unfortunately after two years the window was again opaque. I am going to have to do this again with a better material.