The article presents the results of a multimodal analysis of the covers of British, German, and Polish opinion weeklies, aiming to identify language-image configurations that we term verbal-visual collocations. In our view, verbal-visual collocations give rise to so-called visiotypes, which we define as ubiquitous, one-sided, highly simplifying, standardised visual routines for perceiving reality. These are processed automatically and reflexively, often without the support of verbal cues. They significantly influence and shape the awareness of specific discourse communities. As media-staged, connotation-rich, and highly symbolic images present in the public sphere, visiotypes reflect universal patterns of thought, similar to stereotypes, and are increasingly employed on magazine covers. Since visiotypes represent specific condensed and established combinations of image and text, we utilize a modified version of the model by Stöckl (2011, 2016) and Stöckl and Pflaeging (2022) for their multimodal analysis. Keeping pace with current times, we focus on key figures in international politics: Donald Trump and Vladimir Putin. The analysis addresses questions such as which stereotypical representations and images of these politicians are (re)produced or evoked on the verbal and visual levels and what relationships between verbal and visual codes (possibly metaphorically reinforced) are employed to present the editorial stance on the cover and/or shape and reflect readers’ opinions. Furthermore, one of the goals of this analysis is to examine whether the category of verbal-visual collocation appears adequate and scientifically justified.