Despite the notable progress in realistic image synthesis, there is still a long way to go before one can automatically generate predictable images of organic materials such as plant leaves and human skin. We believe that the predictability of rendering algorithms can be increased by bringing more biological knowledge and data to bear on the modeling of light interaction with these materials. Furthermore, such an approach will likely extend the scope of scientific applications of current computer graphics simulation frameworks. In this presentation, we examine a number of issues related to the improvement and evaluation of such frameworks. We also look at more general open problems, their implications and recent developments in applied tissue optics.