Thanks for your assistance, I think you bring up a good point - the LEDs in the cathode must be throwing off the simulation results, since I don't use LTSpice either, I substituted Rk and Ck in the cathode, and adjusted the Rk for Ia of 8mA. When I do that, both Koren and Ayumi's model gave about the same 2HD. I think when the Koren's model worked better with the LED bias, was just coincidental, since it is really not a good model.
p.s. I will take a look at my old computer to see if I can find some models for the LEDs in LTSpice...