During the workshop at the ISLE conference in Freiburg, the participants agreed to engage in an online discussion on how to bridge the paradigm gap in researching ESL varieties and EFL variants of English. The most productive strand of this discussion concerned the modelling of different Englishes, but some participants also took up the threads initiated by the editors who had selected quotations from the articles in this volume as a possible starting point for discussion. The following is a summary of the main points of the discussion. Some of the threads initiated by the editors are woven into the discussion of the Kachruvian three-circles model in Section 1. Section 2 focuses on the terminological problem of categorizing and labelling features found in New Englishes and Learner Englishes. In Section 3, different developmental trajectories are discussed as the main reason for the ESL-EFL dichotomy. Section 4 addresses issues of corpus methodology and the role of frequency in the description of ESL varieties and EFL variants. The summary of the discussion brings the different contributions together but does not aim at providing answers to questions that can be addressed from different angles. These parallel (and maybe sometimes contradictory?) strands in the discussion might inspire future directions of research.