push(@list, "($1) $2-$3") while /\(?(\d{3})?\)?\s*[-.]*\s*(\d{3})\
+s*[-.]*\s*(\d{4})/g;
[download]

() 209-4044
(440) 248-2700
(440) 975-0537
() 213-0518
(330) 836-9300
(330) 864-5741
(313) 884-6400
(313) 884-0600
(313) 590-9253
[download]

Update: I should explain a little... The regexp breaks down into, basically: optional parens around optional 3 digits, minimal separating junk, 3 digits, minimal separating junk 4 digits. It may look ugly, but it's actually quite straight-forward, although you could obviously tinker with it a little if you wanted. Oh... and the while ... /g means to count it each time it appears on a line (so if multiple phone #'s are on one line, it'll count each one).

For example, I'll turn the "separating garbage" chunk into just [-.\s]*, which is more permissive as well as shorter to write out. Still gets the same results on your sample data.

[me@host]$ perl -ne 'push(@list, "($1) $2-$3") while /\(?(\d{3})?\)?[-
+.\s]*(\d{3})[-.\s]*(\d{4})/g; END{print join("\n",@list)."\n";}' data
+.txt 
() 209-4044
(440) 248-2700
(440) 975-0537
() 213-0518
(330) 836-9300
(330) 864-5741
(313) 884-6400
(313) 884-0600
(313) 590-9253
[me@host]$
[download]

------------
:Wq
Not an editor command: Wq
[download]