This article traces the development of a contextual approach to the teaching of science (physics) subsequently called the Large Context Problem (LCP) approach. This approach is based on the general observation that learning could be well motivated by a context with one unifying central idea capable of capturing the imagination of the students. The design and the placing of the LCP is followed by a detailed discussion of the contexts of inquiry (as we find it in Newtonian physics) that attempts to give the contextual activities found in the LCP a sound theoretical backing. However, the contextual base must also be informed by the findings of constructivist learning theories as well as the history of science. Moreover, thought experiments (TEs) are considered an important aspect of science (physics) teaching and three kinds of TEs are discussed. All these ideas are summarized for a quick overview in a table called "Levels of investigation for scientific inquiry."