Yeah.
You're right.

I would edit that message after you have settled a bit.

"your solution:
split by space and try to decode each "letter"
if next "letter" is '' then it must be word separator - append space, and on next cycle too, and on next
and wrap it in try/except because processing last letter will cause list index out of range. exception in morse[index + 1]"

Not true. Code doesn't iterate trough each letter in every word

Exception neglects KeyError when searching in dictionary

"flow control via exceptions is a big no-no"

Not true for python

Your code (apocryphal , probably)

"for letter in word.split(' ')) for word in morseCode.strip().split(' '))"

Nested loop (efficiency killer)

My code could be improved. What abut "yours"?

did you benchmark and compare this solution to yours?
whats the result?

my guess is you solution is even worse because of regular string reallocations in output += MORSE_CODE[item] and try/except is not free either

double loop solution is simple and easy to understand - split words, split letters, convert, join letters to words, join words to sentence

your solution:
split by space and try to decode each "letter"
if next "letter" is '' then it must be word separator - append space, and on next cycle too, and on next
and wrap it in try/except because processing last letter will cause list index out of range. exception in morse[index + 1]

flow control via exceptions is a big no-no

and it took me almost 10 mins to comprehend your algorithm

You can also start enumerating from any int you want
in this case it would be
enumerate(str(n),p)
I recommend always to check the documentation whenever you see something new