As others have noted, the agreement in effect is horrible.
The fact that 4N is said to be 'often two places to play' simply makes the usage even less viable, since it seems safe to assume that it is 'often three places to play' as well!
Whatever we do is a gamble. Partner will often be passing no matter what. So we may as well aim for the home run, and that is 5
♥.
5
♥ is clearly best when he has a takeout double hand.
5
♥ is clearly best when one of his two places to play is hearts: there are two 2-suiters involving hearts and only one that doesn't.
Admittedly, our shape tells us that the odds of a minor 2-suiter are significantly higher than the a priori 33.3% of 2-suited hands, but while I can't calculate the odds, I still think that hearts as one of his suits, if 2-suited, is going to be about 50%.
If we assume that 50% of the time he has support for all 3 suits and 50% has a 2-suiter, bidding 5
♥ will find our best spot 75% of the time, which is a lot better than bidding 5m.
We can play games with probabilities and I am not claiming any degree of precision but I do think this sort of analysis is useful at the table.
if I were to bid 5m, I'd bid 5
♣ because partner should pull 5
♣on a red 2-suiter. Once is a while we'll have long clubs and nowhere to go when he pulls, but he will almost certainly assume that I would pull to my lowest option when in doubt. IOW, with say 2=4=3=4, I'd bid clubs.
Meanwhile, if I bid diamonds, he should and would assume that my diamonds were better than my clubs...the odds of his pulling to 5
♥ seem to me a lot lower...if he has hearts and clubs he will usually have to pass and hope that my (real-ish) diamonds are really for real
'one of the great markers of the advance of human kindness is the howls you will hear from the Men of God' Johann Hari