The phenomenon of successive language learning acceleration, frequently experienced by polyglots, when in order to learn a new language from a familiar language group the polyglot requires less time with each new language, is widely known, but, it seems, it has never been thoroughly examined. This article presents a simple mathematical model based on the author’s own data, which has been collected over the course of three years’ worth of independent language study and describes how much faster one learns languages from the same group. The number of hours spent on a new language as a function of the number of previously known languages is described by a simple exponential function with two parameters: the “starting time” and the “half-life”. According to the author’s hypotheses, these parameters may provide a numerical measure of certain aspects of language that are difficult to quantify otherwise. The “starting time” could be a measure of propinquity between the learner and the language group, whereas the “halflife” could be a measure of propinquity between the languages of a given group. Additionally, reviewed are three different approaches to keeping track of time spent on language activity as used by different polyglots. These approaches are of importance for collecting data to be used in studies of successive language learning acceleration. At the end of the article, an idealized algorithm for conducting such a study is presented, and particular attention is drawn to the various parameters that must be controlled in order to carry out this kind of research in an appropriate manner. This particular study did not manage to satisfy all of the criteria mentioned, so the reliability of the claims made in this article is debatable, and additional validation is required. Furthermore, the validity of the model has to be confirmed by other researchers and polyglots.