The applied disciplines of architecture and civil engineering require students to communicate multimodally, and to manipulate meaning across media and modes, such as image, writing or moving image. In their disciplinary studies for example, students must be able to transform the language of lectures and textbooks into models and diagrams. In their future workplaces, they will commonly be required to transform reports and legal documents into floor plans and digital & physical 3D models. Such multimodal literacy, however, is not typically reflected in their related subject-specific English language courses, especially in Germany, where a text-centric approach is favored. To better reflect the demands placed upon them, students in two courses of English for Architecture and Civil Engineering were tasked with creating digital, multimodal artifacts to explain a concept from either of these fields to a lay audience. The resultant artifacts used a wide variety of semiotic resources to make meaning, including a total of 26 separate architectural and civil engineering models. This is a quantity sufficiently large enough to invite closer examination, and also reflects the important role models play in the fields of architecture and civil engineering, both at university and in the workplace. This paper suggests that models of this kind exist within a system of signs, in which meaning is created in the relationships between the signs. The process of transforming one resource into another also invites the consideration of the artifacts in terms of the notion of “transduction”, to discern how meaning changes between contexts, practices and modes and to contribute to existing literature on multimodal texts in tertiary education, particularly within a language-learning context.