How does your experimental data in a wind tunnel relate to wing planform? It is a known fact that the elliptical wing planform, as on a spitfire, gives the best spanwise lift distribution, and thus, the most efficient wing design as relating to overall lift and drag performance. I will agree with you about wing tip vortices, but flat wing tips are draggy compared to a sculpted aerodynamic shape. A simple tuft test will show you relative airflow, and that's all. Unless your models were mounted to a balance that was able to read minute differences in lift and drag, your assumption may be a little premature. I will qualify my statements by saying that while all airplanes fly, comaring models to full scale aircraft is like the apples and oranges thing again. It all has to do with reynolds numbers and things that I can't explain in a short reply. I can see how the lack of a wing tip may help in wing rock during harriers and such. It's just a matter of getting dirty tubulent flow on the wing tips earlier, and being able to maintain that condition throughout the maneuver. A more efficient wing tip won't want to give up quite as easily, and may be more prone to the effects of sideslip and other anomalies of the air that we don't normally think about.