The purpose of this chapter is to illustrate the idea behind the notion of mangling as part of the empirical development of didactic models. Didactic models range from macro theories concerned with the selection of goals, content, and methods to micro level modelling of individual lessons and students’ performance, and may take various shapes such as schemata, classification patterns, and rationales for didactic action, i.e., for teaching and learning. Didactic modelling consists of three core activities: extraction, mangling, and exemplification. Mangling designates a process of successive and deliberate adaptation of didactic models by applying them in didactic analysis and design in new contexts. We provide an example of mangling of two existing didactic models, the curriculum emphases (Roberts, Science Education, 66(2), 243–260: 1982) and subject foci (Fensham, Development and dilemmas in science education. The Falmer Press: 1988; Östman, Journal of Curriculum Studies, 28(1), 37–55: 1996). These models were initially developed for didactic analysis of science textbooks and national curricula. In this study we show how these models needed to be modified when mangled through a different practice, viz., analyses of actual teaching in science classrooms. Through the mangling process, the models were modified in the following three senses: (1) Their application range was extended, as a result of demonstrating that they could be meaningfully applied to a new context. (2) The conditions for their application needed to be changed, particular by the need for dividing classroom data into workable pieces that could thereafter be analyzed. (3) The application purpose of the two models changed, from analyses of regularities to analyses of variation in meanings offered to students. These modifications testify to the need for mangling of didactic models in different practices, rather than assuming that they may be unproblematically applied across contexts.