What the exposed compartment adds to SIR
The SEIR model divides a population into Susceptible, Exposed, Infected, and Recovered compartments and tracks the flow between them over time. Its key addition over the simpler SIR model is the Exposed state: a latent period during which a person has been infected but is not yet infectious. For pathogens with a meaningful incubation delay this changes the predicted shape and timing of an outbreak.
Individuals move from Susceptible to Exposed on contact with infectious cases, then become Infected (and infectious) after the latent period, and finally transition to Recovered. The rates governing these transitions encode the latent period, the infectious period, and the transmission rate.
What the model captures and where it simplifies
SEIR is useful for exploring how interventions or a longer incubation period shift the epidemic peak, but it is a deliberate simplification. Basic versions assume a well-mixed population, constant rates, and no age or contact structure, so they describe broad dynamics rather than precise case counts. More elaborate variants add births, deaths, waning immunity, or network structure.
This is general educational information about epidemiological modelling, not medical or policy advice.