he argues that we still need to do it to remember to reduce throttle and flare. Is he right? Would this be acceptable in a real lesson?
If you (for any value of "you") need a reminder for that, then you shouldn't be flying. Seriously. I suppose an instructor could accept needing to call it out or even do it themselves (while the student is following along) the first few times, but reducing power and initiating the flare at the correct point would be something a student would be expected to get the hang of fairly quickly. It doesn't need to be perfect early on, but it should be about right.
Look at it this way. In a small piston-engine airplane, the typical procedure for getting down on the ground after the turn to final is to establish yourself on or maintain an established proper glidepath while maintaining proper lateral position, correcting for wind if necessary; get to a low altitude roughly above but not before the threshold of the runway, with a moderate sink rate; and once you pass the threshold of the runway, flare to reduce to landing speed and arrest the descent, with the aim of touching down at landing speed, with basically zero vertical speed and aligned with the runway.
There's a million variations to this depending on the specifics of the airplane and the runway, and an instructor would help you work out the quirks of handling the particular model you're flying, but that's the gist of it.
To reduce power and flare are absolute cornerstones to getting down on the ground. It's not something that should need a reminder from a second person. It's not like there's a whole lot else going on in the cockpit at the time. (By the time you turn base, you should basically be set up for a landing.)
If you set the wheels on the ground with the engine power in, then at the very least you're going to use up more runway than you need to get stopped. Any good instructor would work with you to break that habit, because sooner or later you're going to be doing short-field landings, or even short-field stop-and-gos. In small airplanes, it's perfectly sufficient, and often encouraged, to fly a significant portion of the landing at a glide. (One instructor I've flown with likes to fly the whole base and final legs at glide, with the engine at idle, and to only push in a little throttle to correct if you end up low as a result.)
A more useful callout in a small airplane would probably be whether you're climbing above or sinking through a proper glidepath, as defined by touchdown on a desired point of the runway with no further glidepath adjustments. (I'm assuming here that your approach speed is well above stall speed and not higher than marginally below a limit speed, which may or may not be the case depending on the aircraft, its load, and configuration at the time; if your approach speed is close to stall speed or close to a limit speed, that's another thing to keep a very close eye on.) However, that's your <> responsibility as pilot, and likely also pilot in command, to judge correctly! Relying on another person in the airplane to make that judgement means you're losing valuable moments before correcting the error, even if that person reaches the same conclusion that you would at the same time you do, because that other person needs to (1) recognize the situation, (2) realize you're not correcting, (3) articulate the need to correct the situation, and (4) wait for you to take corrective action. This should rather be (1) recognize the situation, and (2) take corrective action.
More generally, if you (again for any value of "you") want to simulate flying an Airbus, then simulate flying an Airbus. Don't simulate flying a small single-engine piston propeller plane and pretend it's a commercial jetliner. If your simulator of choice doesn't offer Airbus models and you want to simulate flying an Airbus, get one that does offer Airbus models. (FlightGear is free and open source, hardly horrible, comes precompiled for Windows 7 and newer, and has a large selection of aircraft including the A300, A320 family, A340-600 and A380. I can't vouch for the accuracy of the modelling, though.)