The first, and most visible modification, was the addition of a second set of fins forward of the stock kit fins. This came about from looking at the balsa scraps left from the fin sheets, and thinking that some of the scraps were big enough to make fins on their own. I took the large triangular scraps, and mounted them with the grain along the leading edge, one directly in front of each stock kit fin, with about 1/8" of overlap (that is, the root end of the trailing edge of each added fin was trimmed to butt against about 1/8" of the leading edge of the stock kit fin). All fins, of course, were rounded on leading edges, and tapered thin on trailing edges.
The second mod was to cover the fins with silk tissue (this is a tissue much like the Japanese tissue used on small model airplanes, but made from silk fiber -- it's very strong and light, but the binder doesn't hold when wet, so it requires considerable care in application). This was to prevent splitting and cracking, and had the side benefit of smoothing the fin surface without the added weight of common wood fillers. I applied the tissue with thinned white glue, but if I were doing it again, I'd use nitrate dope thickened with Testor's Extra Strong Cement; it wouldn't "fuzz" the tissue (as much). The tissue covering was applied before attaching the fins to the body tube, so that the root edge of the tissue was bound into the fillets on the joints.
Third, I assembled an ejection baffle from paper towel tube (a loose slip fit into the BT-60 body) and scrap card stock; this was glued into the body tube against the forward centering ring after the motor mount was installed. This baffle worked reasonably well, allowing me to fly with only a single sheet of wadding instead of the 5 to 6 recommended for this kit, but it was destroyed by the "hot" ejection charge the first time I flew an Aerotech composite motor in the rocket. Plywood or balsa next time, I guess.
Finally, I made "spokes" from scraps of balsa -- these were the same thickness as the fins, about 3/8" wide, and long enough to run from the motor tube to the inner surface of the body tube. These were carefully trimmed to fit against the glue fillets and glued to the rear centering ring and motor mount tube, and were then coated with glue along with the centering rings when the motor mount was installed. The spokes were aligned with the fins, so that the body tube at the spokes had solid balsa and glue fillets at the same points on both inner and outer surfaces. These spokes should have made the rocket strong enough to fly on even an Aerotech E30, had the mount been 24mm instead of 18mm. As it was, I never saw any indication of strain on any part when flying on D21 motors having twice the impulse and three and a half times the average thrust as those Estes recommended for the stock Bertha.
BTW, the name Bertha-Winder came from the fact that, with the extra fin set, the rocket had a sort of "air-to-air missile" appearance. It was painted in the stock paint scheme for whenever the kit was made: solid gloss black, and I applied the stock decals, though not in exactly the stock locations (I had to move the body tube decals forward to work around the extra fins). I used Formula U urethane from a spray can for the paint; it covered in one coat, and gave a glossy, "wet" finish with no additional work (other than scraping the mold flash off the nose cone before painting).
This rocket was lost on Flight 12, its second D21-7 flight, March 31, 1996; after a textbook flight, it drifted across a road at the north edge of the flying field and landed about 35 feet up in a large tree. That area is brushy enough that I wasn't able to sling a weighted kite string over the branch or around the shock cord; the tree has a trunk over a foot thick but the branches are too far apart to climb without spurs. By this time, given the amount of rain in this region, I don't expect to find it flyable even if it is recovered.