What you do depends a lot on what equipment/fx you have available, what recording equipment you use, and what instruments you are recording. I could only answer for my own setup. There is nothing "wrong" with putting reverb on individual instruments.
My comment about different "rooms" was just my own preference for when building up a recording on a multitracker.
If you have no effects then recording a track the bathroom is a way to get reverb if you are using a mic to record acoustic instruments or voice. Once recorded, the reverb is always there and can never be removed. Moving the mike nearer or further from the source will reduce/increase the amount of reverb recorded. In this case, the reverb is put on each track. It's not really practical to try to put reverb on the whole final mix in the bathroom.
If you have sufficient effects (reverb and delay/echo) and the recording equipment to do it, I always recommend these FX are placed on afterwards and the sound is recorded initially "dry". (Or as dry as possible within the room you have).
As to what effect to use on what... I agree with the other guys really. Listen to what you think sounds right.
There are some general guidelines. I always say that it's probably better to use a little less than too much. It's very easy to get carried away and overdo the effect. You don't want your recording to be swimming in reverb and echo.
The voice always sounds better with some echo or delay added (in addition to the reverb).
In general, instruments like flutes and violins when playing a melody will benefit from some echo or delay.
As I said in my other post, you need to be careful when putting echo/delay (and in some cases reverb) on rhythmical instruments or tracks.
This is because the delayed sound interacts with the original.
I personally never put echo/delay on a bass guitar, for example, and usually only a little reverb if it seems to need it. The reverb or echo from a bass track can easily swamp the whole mix if you are not careful.
I had a look on the web for sites to explain the delay time and beats per second issue. They all go on about mathematical equations and generally make it even more complicated.
What it boils down to is that the listener will hear a beat, say a snare, twice a second, shall we say. (This is 120 beats per minute, 2 beats per second)
If you put a delay/echo on that, the listener will also hear the echo just after the original sound. How long after depends on the delay time. You need to make sure that the delay time is such that the echo comes before the next beat.
In my example, the echo must come in less than half a second. Most delay/echo units have an adjustment to do this. Just turn it down untill you hear the echo clearly coming in before the next beat. Exactly where the echo happens is up to you. What sounds right.
Reverb is different. The reverb time is a measure of how the reverb sound "decays". In a large hall or cave the reverb lasts a long time before it slowly dies away. In a smaller room the reverb dies away more quickly. Reverb is made up of multiple echoes.
Of course, in a bathroom (not usually very large) the reverb dies down more slowly because the smooth walls create multiple echoes.
Increasing the decay time on a reverb, puts the sound in a larger space.
Increasing the "amount" of reverb (compared to the volume of the original sound) places the sound further back in the mix. It's like moving the mike away from the instrument you are recording and picking up less of the original instrument, and more of the sound from the environment.
I'll stop here because this is starting to sound like a tutorial!
I hope it has been of some help. Remember, use your ears to judge the final sound. But don't overdo the effects!