He really did say that there is evidence of potential innate gender differences (there isn't), and he really did trivialize the possibility (shown in many publications) of the effects of socialization, etc.

If any of these things had been said vis a vis racial differences, the outcry would have been enormous.

The core problem Glass-Stegall addressed was commingling investment banking and brokerage on the one hand with retail finance (i.e., direct to consumer banking services).  It addressed that problem by prohibiting cross-ownership/activity - a company could be either an investment bank or a retail bank/financial service provider, but not both. Moreover, it tried to keep retail brokerage separate from both.

The reasons are, were, and should remain obvious:  

(1) when you can get all the financial services at one stop, the provider of those services has every incentive to structure the transactions you, consumer, have with the provider in such a way as to maximize the profit to that provider.  That profit comes from the consumer's wallet.
(2) the risks one side of the provider's business takes are inevitably communicated throughout the entire business of the provider.  This is an example of something lawyers call "cross-collateralization".  To oversimplify an example, in the context of an individual, a cross-collateralization would be putting your car up as security for the loan you are taking out to buy that big-screen TV.  Everything is fine as long as you keep making the payments on everything (i.e., the income stream continues) but the minute the model breaks down (an unexpected expense or interruption in the income stream), everything collapses in a hurry.

Sound familiar?  Glass-Stegall was enacted as an empirical fix to the markets, to prevent precisely the same kind of collapse-spreading we've just seen - which was behind the 1929 crash.  In other words, by forcing companies to stay in only one market segment, it both provided a firebreak against collapse, and forced realistic appraisal of risk.  Risk could not be palmed off from an unsafe part of a company to a safer part of that company.

AIG, for one, and Bear and the other investment houses all had lots of profitable parts of their overall business.  But they wound up in or on the way to the graveyard because they took risks in one part of their business which risks were communicated throughout the whole of their business.  While a lot of people - to a degree correctly - have attributed the rapid spread of the collapse to the extent of leverage in mortgage derivative and other transactions, it was the "all under one roof" model of the finance company that all these businesses subscribed to that led inevitably to the whole edifice collapsing.  And it was that "all under one roof" model that Glass-Stegall prohibited.

I recall a smaller, though no less significant, incident back in the 90s - during the height of the tech/dot-com boom.  I was dating someone who worked in one department of a large investment house.  Christmas bonus time came and top management announced "no bonuses this year".  And layoffs, right before Christmas.  This was not because they had done a bad job, nor because they had not made profit.  It was the dot-com boom and they were printing money and picking winning stocks by tossing darts at the agate pages of the WSJ. Rather, it was because a couple knuckleheads trading derivatives had made a series of trades which wiped out the entire company's profit for the year, from which bonuses would have been paif.  "Knuckleheads" was a post-hoc name - they would have been heroes had the trades been profitable.  But, they weren't.  And everyone suffered.

The CDOs and mortgage-backed securities which are the kernel of the issues we're presently dealing with are, in their essence, a form of insurance.  They spread risk.  But, the companies putting them together were allowed to escape all the regulations (not the least of which are loss reserve requirements) pertaining to insurance, merely by saying they weren't insurance.  In other words, even though it had webbed feet and feathers, walked like a duck and talked like a duck, by calling it a pony, the companies were permitted to put a saddle on it and ride.  And sell tickets to ride.  As long as that pony's stubby little legs and webbed feet held up, it was all giddyap all the time.  And everyone believed it was a pony.  

But, when that pony's little legs gave out, the same dunderheads waiting in line stood around saying "why did you call this duck a pony?  I was supposed to ride it next - I've already paid for my ticket."  And the ride operator was already over the hill with the gate proceeds.

The thing about insurance - and CDOs, swaps and all the rest are insurance, regardless of being called something else - is that it is a boring industry.  The actuarial tables tell, in painful detail, how much the risk is.  And they prescribe how to address risk.  The laws and regulations governing insurance pretty much prescribe how much profit can be made on it.

All of this is pretty much anathema to your typical investment hotshot who wants more than the law allows - for himself.  And needs the rush of beating his buddies in the number of dollars' profit he puts up on the scoreboard.

But most importantly, every time an insurance business fails, it fails because someone in a senior position in the company tried to get around the rules and tried to make more profit than the laws and regulations provided.  You can look to the history of insurance company failures and you'll see "overinvestment in junk bonds", "inadequate loss reserves", "overinvestment in penny stocks", "failure to diversify investment risk" and similar causes in every case.