The paper is a continuation of the author’s preceeding article Constructivism as a Philosophy and a Sociology of Science (Filozofia, 63, 2008, No 1). It sheds light on Goodman’s irrealism as well as on the place of the constructions in producing the plurality of worlds. It gives the definition of the strong constructivism, which rejects the idea of representation, while one-sidedly stressing the external social determinants of knowledge (K. Knorr-Cetina, the social epistemology of S. Fuller etc.). In particular, some properties of the constructs are shown as related to the information and its processing. In conclusion some weak versions of constructivism as the expressions of an exaggerated reflexivity are questioned.