Many important aspects of cancer biology, such as cancer initiation, progression, and metastasis, have been studied in animal models, mostly mice. As long as cancer was considered primarily a genetic disease, the study of transplantable mouse tumors, or even human tumor xenografts in immunocompromised mice, appeared to suffice. Many important genetic events that lead to transformation and in vivo tumor growth were elucidated. However, many even more important factors that determine whether or not the genetic potential of a tumor cell will be realized, such as the host response to the tumor and the tumor microenvironment that influences this response over a long period of time of tumor development, remained untested and unappreciated. This is slowly changing with the advent of molecular techniques that have spurred efforts to engineer better mouse models of human tumors. In this review, we show results of our efforts to combine a genetic mouse model of spontaneous human adenocarcinomas based on a Kras mutation, with an important human molecule MUC1 that is abnormally expressed on human adenocarcinomas, promoting oncogenesis, proinflammatory tumor microenvironment, and immunosurveillance.