The use of information granularity is regarded as an essential design asset whose optimal allocation helps in augmenting existing models. This chapter discusses several development scenarios and classes of numeric models in which the allocation of information granularity provides tangible benefits in order to better capture the experimental data, form granular conclusions, or offer a more detailed insight into the parameters of the original models. Local sources of knowledge structured in the form of logic descriptors—constructs of fuzzy logic, are arranged together in the form of a global model coming as a high-level granular logic descriptor. In a nutshell, knowledge transfer is about forming ways that an existing source of knowledge can be used in the presence of new, very limited, experimental evidence. The chapter deals with a collection of logic descriptors describing some local logic; pieces of knowledge are arranged together to form a global description of available knowledge.