Short pulses of a probe laser have been used in the past to measure wh
ether a two-level atom is in its ground or excited state. The probe pu
lse couples the ground state to a third, auxiliary, level of the atom.
The occurrence or absence of resonance fluorescence were taken to mea
n that the atom was found in its ground or excited state, respectively
. In this paper we investigate to what extent this procedure results i
n an effective measurement to which the projection postulate can be ap
plied, at least approximately. We discuss in detail the complications
arising from an additional time development of the two-level system pr
oper during a probe pulse. We extend our previous results for weak pro
be pulses to the general case and show that one can model an ideal (pr
ojection-postulate) measurement much better with a strong than a weak
probe pulse. In an application to the quantum Zeno effect we calculate
the slow-down of the atomic time development under n repeated probe p
ulse measurements and determine the corrections compared to the case o
f n ideal measurements.