How is the Mira Cortisol different from serum or a 4-point salivary cortisol?Updated 3 hours ago

Serum measures total cortisol (about 90% protein-bound) at a single moment, and is strongly time-of-day dependent. It is the standard sample for a morning cortisol in a suspected adrenal insufficiency workup and for stimulation and suppression testing. 
Urinary free cortisol reflects the unbound fraction excreted over the preceding hours, so it is integrated and lagged rather than instantaneous. Because it is filtered from the free fraction, it is less influenced by cortisol-binding globulin than total serum cortisol. A "normal" serum cortisol and an unusual Mira pattern are not contradictory; they are different quantities on different timescales. 
Neither the Mira score nor a Mira pattern should be read as a proxy for a serum value.

Saliva reflects free cortisol at the moment of collection. That is what makes it possible to measure the cortisol awakening response (the rise 30 to 45 minutes after waking) and the diurnal slope, and late-night salivary cortisol is a guideline-endorsed screen for Cushing syndrome. 
Urine cannot capture the awakening response at all: a first-morning void reflects the overnight period, and a second void an hour later averages the whole rise-and-fall into one accumulated sample. Mira gives a coarser view: overnight and, optionally, evening, compared against the person's own history. It is not a diurnal curve in the clinical sense.