In the past decade, in vivo models for optogenetic applications have gained importance, especially in the fields of cardiology and neuroscience. To reliably evoke the desired reactions while minimizing adverse effects, the stimulation power must be carefully adjusted. The relevant light intensity for cells in deeper layers of scattering tissue is not easily extrapolated from the power threshold of single cells. In this study, we evaluated a model for deep tissue optogenetic stimulation, using a heart-like cell line and tissue phantoms. Phantoms were fabricated from PDMS and titanium dioxide particles and possessed highly reproducible optical properties. Scattering and absorption coefficients were modeled to match those of realistic tissues. Since power of light traveling through tissues decays exponentially with respect to the scattering and absorption coefficients, the required input power was expected to increase exponentially by the same factor. To test this hypothesis, cells were stimulated through tissue phantoms of varying thickness with different modes of illumination. Cellular reactions revealed that the simplified assumptions were not sufficient to predict the input power required to reach the stimulation threshold. We provide a more comprehensive model to assess cellular reactions in scattering tissues a priori. This study has implications for the use of optogenetics in tissue models, organs and in vivo models as the outcomes can be transferred to different types of cells and tissues.