One thing I forgot to mention is that the PSR-E433 has a maximum polyphony of only 32 notes, although that's a bit confusing/misleading because the actual polyphony varies depending on which voices you're using.
The PSR-E433 has 32 tone generators, which is where the "32 note polyphony" comes from. But Yamaha uses "elements" to create their voices, which (if I understand correctly) are like layers of sound-- i.e., each element is a sound played by a tone generator, so if a given voice is formed from, say, four elements then each note played with that voice actually uses four tone generators rather than just one.
The rule of thumb seems to be that the maximum number of elements used for any given voice should not exceed the total number of tone generators divided by 16 (because there are 16 MIDI channels). Thus, keyboard models with larger maximum polyphonies-- 128, 192, 256, etc.-- can have voices formed from more elements.
The voices on the PSR-E433 are formed from either one or two elements, so that means the maximum number of simultaneous notes you can play with a given voice is either 32 or 16, depending on how many elements the voice has.
You must also keep in mind that layering a Dual Voice with a Main Voice will also use up some of the available tone generators. And any notes which are sustained-- either by holding down their keys or by depressing the sustain pedal-- will also use up the tone generators.
The reason I'm mentioning all of this is because it can have an impact when you're playing the keyboard in unison with a MIDI song. What generally happens is that if the keyboard's maximum polyphony is exceeded then you'll start getting "dropped notes," meaning notes that get cut off prematurely. Usually a keyboard will drop the oldest notes first.
If you find yourself running into this situation, there are a couple of things you can do to help:
(1) You can open the MIDI song file in a MIDI editor on your computer to identify the voices used on each MIDI channel. Then you can "revoice" the song so it uses only voices formed by a single element. Yamaha used to show the number of elements for each voice in their keyboard manuals, but they seem to have stopped doing this. You can still find copies of the old XG specifications online, and the voice lists in those specifications show how many elements are used for each XG voice. But another way to try to figure this out yourself is to play the lowest key on your keyboard as loudly as you can-- using a sustain pedal to keep the note held-- then quickly (and quietly) start running down the keyboard from the highest key (still sustaining the keys) until you hear the low note get dropped. (That's why I suggest playing the high keys quietly-- so you can still easily hear the sustained low note.) If the low note gets dropped after playing 16 high notes then the voice uses two elements. But if the low note doesn't get dropped until after you've played 32 high notes then the voice uses one element. By the way, you might want to revoice your MIDI song files even if you don't plan to play along with them, because they might have been created for a keyboard with different voices than the PSR-E433, or that has a higher maximum polyphony-- so it's always a good idea to "optimize" a MIDI song file or style file for whichever keyboard model you happen to have.
(2) Another thing you can do is turn off one or more of the song tracks. This is especially helpful if you want to play the part that's on a given channel. The PSR-E433 has six track buttons that can toggle the song channels or style channels on and off. There are (usually) more than six channels in a song, so you can't toggle all of them on and off, but hopefully the channel you want to turn off can be toggled using one of the track buttons. If not, you could always load the song file into a MIDI editor to remove the desired track(s)-- just be sure you keep a copy of the original file.