Another point is that we didn't worry about heterosexuality in this model. In a full model there would be homosexual, heterosexual, and bisexual individuals of each gender, with the corresponding constraints about which partnerships they could enter ...
Key difference is boxes for couples, which generates many parameters for describing partnering and un-partnering.
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1005453
Related to #207
From @bbolker:
Another much simpler model from @wzmli https://www.sciencedirect.com/science/article/abs/pii/S0140673608616979?casa_token=Hf475UmwVjEAAAAA:Ikev2Lkff2ZHy3AB7X8xfUsx6FVIJ4FUN__q74qJ-jB2R-Ze1Ev7s-xOF3KKOnA0Ny4eJtv_np4