We've written several papers on modelling near-IR images of protostars, concentrating mostly on the Taurus molecular cloud, where we have a lot of information (data) to work with. Our first papers were on the Monte Carlo methods we use (Whitney & Hartmann 1992, Whitney & Hartmann 1993). Then we modelled the fluxes and images of the Taurus protostars (Kenyon et al. 1993). These models predicted very large polarizations which we later confirmed with observations (Whitney, Kenyon, & Gomez 1997). That was very satisfying, to make a prediction and then observe it. Our models were able to fit the fluxes of the protostars when we included bipolar cavities in the envelope to allow more optical/near-IR light to escape. Partly because of this, Mercedes Gomez and I directed our jets survey to concentrate on the protostars instead of the more evolved T Tauri stars. This proved very fruitful. This whole series of projects was a lot of fun.
Now that the Hubble Space Telescope has gotten such a "close-up" view of many of these sources, we are challenged to try to understand their complicated geometries. I've spent the last few months updating my Monte Carlo codes to provide much higher signal-to-noise and to deal with 3-D geometries. Soon we'll begin modelling these sources in earnest. Meanwhile, here are some preliminary results: