Every tissue has its own T1 and T2. What actually shows up as bright or dark in the image depends on TR (repetition time) and TE (echo time) — the two timing choices the operator makes. Drag TR and TE and watch the same four tissues flip between T1-, T2-, and proton-density-weighted contrast.
Signal model (spin-echo): S = M0 · (1−e−TR/T1) · e−TE/T2. Swatch brightness is each tissue's S, normalized to the brightest tissue at the current TR/TE.
Try this: set TR short (~500) and TE short (~15) — fat is brightest, CSF is darkest: T1-weighted. Now push TR long (~4000) and TE long (~100) — CSF becomes brightest: T2-weighted. With TR long and TE short, the tissues mostly separate by how much hydrogen they contain rather than by T1 or T2: proton-density weighted.