This article investigates how the creators of Viz use media and modes to produce potentially humorous constellations. The aim is to identify novel patterns of media/mode configurations in humour. For this purpose, theoretical categories are developed which reflect various constellations of media, modes and localities in the production of Viz humour. These categories go beyond the well-researched combination of (written) text and image and range from merely humour-enabling constellations (Category 1), via various combinations of modes to introduce incongruities (Categories 2–4) to two different types of constellations that not only carry incongruities but are incongruous in themselves (Category 5). The investigation shows that in Viz, incongruities are most commonly carried by text and image simultaneously. In addition, media/mode constellations are used which so far have been under-researched: the suggested but never realised physical manifestation of objects or manipulations of the temporal dimension of discourse.