The fast and dense winds of massive-star binaries produce thermal X-rays via shocks at their wind-wind collision regions. The X-ray emission and absorption are both sensitive to the wind mass-loss rates, wind speeds and accelerations, and stellar separation (among other properties), though they each manifest differently. For example, X-ray emission is proportional to density squared while absorption is proportional to density, and absorption affects only the soft X-ray spectrum. Therefore, it is useful to construct dynamic models of a binary’s X-ray emitting and absorbing gas in order to extract as much information as possible from its X-ray observations. This work constructs 3D hydrodynamic simulations of the colliding winds in several systems with well-studied X-ray observations, e.g. eta Carinae and WR 140, and then synthesizes the model thermal X-ray emission from the hydrodynamic simulations, while accounting for wind absorption, to generate phase-dependent X-ray light curves, spectra, and line profiles. Folding the model X-ray observations through the X-ray detector response functions then gives a one-to-one comparison between the models and observations, where discrepancies are used to update the stellar, wind, and orbital parameters of the massive-star binary under study. This presentation will highlight recent results, including the X-ray minimum of eta Carinae lasting the appropriate 2-3 months, though with a need to delay periastron relative to other diagnostics; the good agreement between X-ray line profiles of eta Carinae on the approach to periastron, which is important since this diagnostic also includes the line-of-sight velocity information of the X-ray emitting gas; and the well-matched model WR 140 light curve, allowing for fine tuning of the wind parameters of the system.