Turing machines are part of the prediction-generating framework of CS, like how the mathematical models in the Standard Model are part of the prediction-generat

Turing machines are part of the prediction-generating framework of CS, like how the mathematical models in the Standard Model are part of the prediction-generating framework of modern physics: They're both an idealized representation of something which we think models something useful in the physical world in a way that allows us to think more clearly about some aspects of that thing without getting caught up in distractions. As alex_smart says, the Church-Turing thesis connects the idealized Turing machine to modern computers in a way somewhat similar to how the Standard Model connects the U(1) group to the study of electromagnetism.