whatever you do proceed with your best work first - if you don't know that order, mock it up and get feedback from someone critical you trust

I agree - the best to the Front, hopefully you won't get to show that other work you have included to demonstrate your diverse talents in other media forms. This is Architecture right?

best in front is a good idea, but make sure to have a strong finish.

alternatively, think about the "story you are telling".
assuming that you are not including any "filler" pieces, all your work should be strong.
try to organize in a logical fashion that allows you to tell a story about yourself or your work...does it all fall into a theme that you would like to carry on through your professional career (interest in affordable housing or homelessness.)

remember that the firms are generally looking for 2 things when looking at your portfolio. 1) the work itself and 2) your ability to communicate the work to them.

this communication should not be in the form of verbiose text.

i agree with grace kim, but for my own porfolio i chose the best works from all five years of college and organized those from least to best. the thinking is that all of the projects in my folio are of high quality, so whomever is reviewing it will be impressed from the start. at the end is the "grande finale", so to speak. in short(too late) whatever you choose to do just make sure it flows and shows your desired intent or tells something about yourself. good luck

i submitted two different portfolios to grad schools, because i applied during my undergrad and then decided to take some time off. i strategized my first portfolio in accordance with quality; i.e. i put my best projects at the beginning and end, and a good one in the middle. this supposedly allows people to be drawn into it as well as leave it with somewhat of a good impression. it was successfull, i got into the school i wanted to with a substantial grant. but later i applied to some different schools, allocating one single project to a 12 page f/b portfolio. there were enough explorations into different facets of the project to allow me to do this (and the work overall was a few notches above the rest of my projects) so if you have one really great project you might try that. it doesn't matter what you put in, just save the weak stuff.