Point: You are on a warship with warship sensors. I doubt any commercial submarine would have the kind of gear you had of that era (or that many people devoted to sensor data processing).
Second Point: Arrays used in space need to be tuned and in some cases, your choice is: Faster sweep, lower resolution of data. Longer sweep, greater resolution of data. So, if you calibrate for a sweep (or sweeps) that will take a week and cover a lot of space (a system or even a world you are looking at light years away) or you can calibrate for a longer, slower set of sweeps that spend more time focusing on each particular pixel - kind of like scanning a photo on low res versus high. Yes, the sensor is the same, but the data recorded is not because the hardware's behaviour is actually determined by the parameters it is configured with and that does change the resolution of the data. You never *have* a perfect bit of raw data (earthly example: analog photos using film) but instead have what you sampled from your sensor configured for a certain pace of tracking across space and a certain focus (example scanning at 320 px/in or scanning at 1080 pix/in).
I say this from gamed and discussed with several people working for NASA or affiliated companies, including one who was an infrared imaging expert. Now, he did say, as a counter-point, when you have enough horsepower in computers and in scan gear to run a full sphere scan to your highest resolution in a snappy time, you've got the best of both worlds.
However, now, and at least for another 40, 50, 75, 100 more years, we are not likely to be able to do this sort of 'peering at foreign systems and making out all man-made objects plus some pretty spiffy details about the planets and sun). Now, by the time PA:VT happened, if you are talking TL-10+, you might well be able to run such a scan in one pass. (Note though that the interpretation Alex had is more in line with the feel of Traveller as having kinda weak compute resources...)
Either your interpretation or Alex's could be reasonable depending on how you want your setting to operate.
I prefer the multi-stage scan.
There's also another logic for no single scan model - If you could always get all the data at the resolution you need then analyze it later, it would be what everyone always does and the focus would be on fast computers so that could be done in near realtime (or at least really short time). If it does matter how you setup your sensors which can limit the amount of meaningful data you can record.
The reason I prefer this is because, as a GM, it offers players an important strategy choice that can impact what they know. I like those sorts of dilemmas and I like the choices to have differing results.
But you interpretation could certainly sell if you thought computers were good enough and sensors able to pull in large amounts of data at high resolution with fast sweep rates.
TomB
We decided we'd buy hand-me-downs from the Brits, let them sit in harbour there unmaintained for a while, and then finally pay for them, realize their torpedos were no longer being manufactured, and somebody let a wave crash in through open hatches and start a dangerous electrical fire on one that put it out of commision for a long while.
We really need to improve our procurement planning - a little at a time consistently would be better than a bigger order every 25 years....